If the numbers are not the same how can you compare? Not only that but these are totally two different tests.
Ultimately, we want to compare the Nikon fl to the Sigma Sport. Ephotozine has tested the Canon III and the Sigma Sport on the 5DSR, and the Nikon fl on the D810. We want to figure out what the Nikon numbers would be if tested on the 5DSR, so that we can compare it to the Sigma Sport.
Lensrentals has tested the Canon II and Nikon fl on their optical bench, thus their values are comparable. Assuming the Canon II is as sharp as the III (as most seem to find), we can thus get an objective comparison of how the two compare. If we can determine the difference between them, we can multiply the Ephotozine Canon results by that difference to get an idea of how the Nikon would perform.
I choose to look at the 30 lp/mm results, as this is common in MTF tests. However, it may not be the best choice. Regardless, it gives us an idea that if we use this approach that it would show the Nikon as having sharper edges than the center, which doesn't seem right.
Of course, this is obvious when you look at Ephotozine's coverage of the Canon vs Lenrentals. Clearly something's wrong - Ephotozine shows the Canon as having edges almost as good as the center, while Lensrentals shows a more significant dropoff at the edges.
Either Ephotozine has a very unique lens, one of the two sites is testing things wrong, or Canon does something special with their sensors that makes edges sharper (or the center less sharp).
So ultimately, the exercise failed and I couldn't find a good way to compare the Nikon to the Sigma.
